Unsupervised Learning for the Comprehensive Exploration of Continuous-DAS Data

Distributed Acoustic Sensing (DAS) enables cost-effective, dense detection of seismic activity. However, the vast amount of data produced by DAS systems presents a significant challenge for labeling and analysis. Traditional supervised machine learning approaches require extensive labeling, which is time-consuming and prone to user bias. To address this, we propose a two-step processing chain. The first step constructs a latent data representation from several hundred features. We compare two approaches: one using signal processing metrics commonly used in seismology (human-engineered features) and the other using self-supervised learning with common DAS data representations (image-BYOL). The second step applies unsupervised clustering to reduce the dataset. We first apply K-means to obtain 5000 clusters, followed by hierarchical clustering, merging them into 500–700 clusters using an inconsistency criterion. This method is applied to DAS data from two experiments in the Pyrenees with different configurations: a 6-week measurement on an 800-m cable at Viella and 19 ten-minute measurements on a 91-km cable. For Viella, we detected all events with Mw >2.0. The image-BYOL approach produces more false positives than human-engineered features. The results highlight clustering’s potential for DAS analysis while emphasizing the need to reduce false positives, especially for smaller seismic events.